We should no publish a ft index

[kelson42]

We don't have and need a ft index right?

_ftindex:false as tag should be created

[benoit74]

We do have a full-text index ATM. Content was not that helpful, but it did not really hurt either. I'm fine with the idea of removing the full-text index for the time being.

[benoit74]

Note that _ftindex:yes is already present and aligned with ZIM content, see e.g. https://dev.library.kiwix.org/raw/maps_en_monaco_2026-03/meta/Tags

> curl https://dev.library.kiwix.org/raw/maps_en_monaco_2026-03/meta/Tags
_sw:no;_ftindex:yes;_pictures:yes;_videos:no;_details:yes

[kelson42]

What kind of service provides the ftindex? To me it delivers always no result.

[benoit74]

What kind of service provides the ftindex? To me it delivers always no result.

It does deliver same result as suggestion index.

[kelson42]

We should remove it and update _ftindex tag accordingly

[rgaudin]

We should remove it and update _ftindex tag accordingly

I think it's more practical and reasonable to only set _ftindex:yes if there's a FT index and simply not set it otherwise. I thought that libzim or scraperlib was automatically adding it when using configIndexing() but apparently not… @benoit74 WDYT?

[kelson42]

We should remove it and update _ftindex tag accordingly

I think it's more practical and reasonable to only set _ftindex:yes if there's a FT index and simply not set it otherwise.

AFAIK, this is clearly not what has been decided and written in the spec years ago, this tags have to be explicit. Otherwise it means "undefined".

[benoit74]

This is a fork of the core topic so I propose to move this discussion to openzim/overview#82 and keep this issue focused on removing the ft index from maps ZIM. But thank you for speaking up, I indeed recall about that discussion about not setting the tag just to say "no".